Endometrial cancer is the most common gynecologic malignancy in US women, with more than 39,000 new cases each year. Stage II or higher endometrial cancers recur in more than 30% of cases, and most recurrent cases result in death. However, little is known regarding the biologic factors that underlie endometrial cancer recurrence. We reason that many of the same risk factors associated with the initial development of endometrial cancer may also represent risk factors for recurrence (e.g., by stimulating the growth of residual tumor and pre-cancerous cells). Two of the strongest risk factors for endometrioid adenocarcinoma (EA), the most common form of endometrial cancer, are estrogen exposure and obesity. In fact, high estrogen levels are common in obese postmenopausal women, and are thought to partly explain the obesity-EA relationship. Hyperinsulinemia is also common in obese patients, and our recent prospective study in the Women's Health Initiative (WHI) found that insulin - which has mitogenic/anti-apoptotic activity - was significantly associated with risk of incident EA (hazard ratio, highest to lowest quartile = 2.33; 1.13-4.82), after controlling for estradiol. We also studied insulin-like growth factor-I (IGF-I). IGF-I shares sequence homology with insulin and is mitogenic. However, uterine and serum IGF-I levels are regulated by different mechanisms, and it is not known whether they are at all correlated. It is notable, therefore, that we found an inverse relation of EA with free IGF- I levels in serum, results that are in keeping with most prior cross-sectional studies, and could reflect the known anti-inflammatory activity of IGF-I in serum. An indirect effect such as this is consistent with there being little correlation between serum and uterine IGF-I levels or activity, a possibility that requires further study. Determining the role of insulin, the IGF-axis and sex hormones in EA recurrence has important clinical implications. First, it could provide a means of differentiating EA into prognostic subcategories (e.g., to triage patients into more or less aggressive therapy), and second, improved understanding of these tumorigenic pathways could lead to novel therapeutic approaches. Therefore, in collaboration with the Gynecologic Oncology Group (GOG), we propose the first large prospective study of the insulin, IGF and sex hormone axes and their associations with recurrence of stage II-IV EA. Subjects will include all of the 815 stage II-IV EA expected to be enrolled in GOG-0210, a study of endometrial cancer patients enrolled prior to treatment. More than 170 of the stage II-IV EA cases are expected to recur. Pretreatment specimens will be used to address the following aims. To study the associations of EA recurrence with: (i) fasting serum levels of insulin, total and free IGF-I, IGF-II, IGFBP-1 and -3, estradiol, estrone, progesterone and sex hormone binding globulin (SHBG); (ii) tumor expression of IGF-I, IGF-II, IGFBP-1 and -3 mRNA (based on RT-PCR), and of the insulin receptor, IGF-I receptor, estrogen receptor, progesterone receptor (based on immuno-histochemistry). Lastly, we will (iii) measure the correlations of serum and tissue IGF-I, IGF-II, IGFBP-1 and -3 in women with and without cancer. PUBLIC HEALTH RELEVANCE: Endometrial cancer is the most common gynecologic malignancy in women, and higher stage endometrial cancer is associated with high rates of recurrence and subsequent death. The proposed application will investigate the prognostic value of both serologic and tissue levels of sex hormones and insulin/IGF axis components in relation to endometrial cancer recurrence. The study will be conducted in a large, prospective cohort study of endometrial cancer (GOG-0210).